Data rich, insight poor

How to use data-science to empower management and conservation

Get the slides

Data-science means using data for responsible decision making. It has been embraced by companies in many fields for optimizing their processes and improving management resources.

Download .ppt »

See it in action

See how real-world monitoring data can be explored on the fly. Data was collected as part of the subsistence fishing monitoring program by the Seychelles Islands Foundation

View dashboard demo »

Try it yourself

This demostration takes data from an online/offline electronic survey that you can fill yourself. Then, analyses the collected data can take place in real-time to enable quick decision making.

Real-time demo » Fill survey »

Make your own

Follow a step-by-step guide to create your own electronic survey and real-time dashboard using KoBoToolbox, Qlik and some of the basic tools of the R package Shiny

Tutorial »

Sign up for a data collection tool

For this tutorial we will use KoBoToolbox, a free and open source set of tools for field data collection. It helps you to create surveys that can be conducted entirely while being offline, allowing its use in challenging environments. One of the crucial features of KoBoToolbox is that is easy to use, and with very little effort you can set up an account and design your own survey.

There are many other data collection tools available though. Ona, Enketo and Open Data Kit, and ProofSade are some of them.

Create Account »

Create your first electronic survey

Once you've successfully logged in to KoBoToolbox, you can create your first survey from scratch by clicking the NEW icon and selecting Project. Choose a Project Name and a short description for your Project. This will create an empty form where you can add any question you want.

Clicking the icon will add a new question. For the purpose of this tutorial, we will keep this form as simple as possible by simply adding a question named 'Location'. Select the Point option to create a map for the users to add locations. Use the to select a 'Mandatory Response' in order to avoid empty entries. When creating real world surveys, make sure that the field Data Column Name is a short word easy to remember.

Save the form by clicking SAVE and finish the edition with the icon. By doing this, your form will automatically be stored as a draft in your account. Check what the form looks like by clicking the icon. If you are happy with the results, you can select DEPLOY to create your electronic survey. Otherwise, you can edit the form as many times as you need.

If you now check select the Deployed tab you will find the current version of your electronic survey. Notice that you can edit and redeploy the survey if necessary. Try to fill out the form multiple times as if you were collecting data in the field. To do so, you can open the form by clicking the OPEN option in the 'Collect data' section. Alternatively, you can ask somebody to fill out your form by sending them a link (use the COPY option to obtain the link to the form). This will generate multiple entries in your database.


A quick exploration of the data

There are many ways to quickly explore the results of your survey. One option is to directly explore your data using KoboToolbox to have a first look at it. Try selecting the DATA tab on top of the page and start exploring your data.

However, usually you might want to other more useful tools. Platforms such as tableau and Qlik provide you with a friendly interface to analyse and explore your data. These tools are used by many small and medium companies to provide business intelligence on the fly. Data can be downloaded from KoboToolbox by simply clicking the Downloads tab and selecting the file format that you need.

More powerful analysis can be made using programing languages such as R or Python. If you are familiar with them, you can even interact with your data without downloading, ensuring that your script always analyse the most updated version of your data. In R, for example, few lines of code will automatically import your data to your R session (see the code attached). Try downloading the code and adding your username, password and the id of your form. You can find the id of your forms here

  link <- paste("",
             as.character(id), sep="/")

  data <- httr::GET(link,
             httr::authenticate(username, password))

  data <- httr::content(data)

  data <- jsonlite::toJSON(data)

  data <- jsonlite::fromJSON(data)

  data <- strsplit(unlist(data[,"Location"]), split=" ")

Download R code »